An improved multi-objective honey badger algorithm based on global searching strategy

This paper proposes an improved version of the Honey Badger Algorithm (HBA) for multi-objective optimization problems, referred to as the Improved Multi-Objective Honey Badger Algorithm (IMOHBA). The collective behavior search strategy of the HBA is integrated with a dynamic archive to efficiently retrieve and store Pareto optimal solutions. Additionally, a leader selection mechanism based on crowding distance and the roulette wheel strategy is employed to select the optimal solution in multi-objective space. To overcome the issue of local optima, a modified mutualism phase from the Symbiotic Organisms Search (SOS) algorithm is introduced to enhance the global search capability. The algorithm is tested on CEC2009 benchmark functions and various real-world engineering problems, demonstrating competitive performance for solving complex multi-objective optimization problems.